fixed problem in range_to when hue should've cycled.

[alexhagen]

Previously, when performing range_to on colors with hues that crossed 1.0, the color_scale function used in range_to would go through all intervening hues instead of crossing 1.0. For example, when going from purple with hue 0.94 to red with hue 0.04, range_to returned almost all colors (blue, green, yellow) by ranging from 0.94 downward to 0.04 instead of going from 0.94 to 1.0 and then restarting at 0.0 and going to 0.4. This commit fixes that.

[codecov]

Codecov Report

Merging #48 into master will decrease coverage by 2.32%.
The diff coverage is 88%.

@@            Coverage Diff             @@
##           master      #48      +/-   ##
==========================================
- Coverage   99.13%   96.81%   -2.33%     
==========================================
  Files           1        1              
  Lines         232      251      +19     
==========================================
+ Hits          230      243      +13     
- Misses          2        8       +6

Impacted Files	Coverage Δ
colour.py	96.81% <88%> (-2.33%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 11f138e...8a6a0cb. Read the comment docs.

[vokimon]

Shouldn''t be simpler, in case we detect abs(huea-hueb)>0.5, just adding 1 to the lesser hue and modding resulting hues by one?